VUB Artificial Intelligence Lab demonstrates importance of artificial intelligence in the fight against epidemics
New VUB research shows that artificial intelligence (AI) can successfully support policy makers in the fight against epidemics such as COVID-19. For this purpose, it combines epidemiological models with complex decision-making models capable of taking into account factors that are difficult to predict, such as human behaviour. This research is one of the many initiatives of VUB scientists to deploy cross-disciplinary AI expertise in the fight against COVID-19.
In his research, Pieter Libin studies epidemiological applications of Reinforcement Learning, an AI technique that became famous when Google Deepmind's AlphaGo programme defeated the world champion in the GO game in 2017. In this form of AI, well studied mathematical techniques and concepts derived from the domain of psychology are used to make optimal decisions based on interaction with the environment. In addition, Reinforcement Learning is particularly suitable for mapping complex interactions and thus involves behavioural, cultural and social factors in determining successful prevention strategies to achieve well-specified goals when tackling an epidemic.
Tackling epidemics more efficiently with AI
Pieter Libin's PhD* from the Artificial Intelligence Lab (VUB) concludes that the use of artificial intelligence better enables policy makers to tackle epidemics and pandemics efficiently. Artificial intelligence allows to propose a very targeted approach in the fight against epidemics. The models are capable of learning an optimal strategy, which can consist of closing schools, making it compulsory to wear mouth masks in certain activities or vaccinating certain target groups, provided a vaccine exists.
In addition to Reinforcement Learning, other machine learning techniques are also useful in the fight against epidemics. Combined with statistical and epidemiological models, they make it possible to estimate how quickly the epidemic will spread over time. This in turn makes it possible to determine the number of intensive care units that will be needed in hospitals. In this way, medical institutions are better able to make the necessary adjustments to their infrastructure in advance. Artificial intelligence can also play a major role in the development of medicines. For example, pattern recognition techniques can extract active compounds from bioinformatics databases to contribute to the design of a new drug against the virus.
"Our research shows that AI offers promising opportunities to support policy makers in their decision making and that it can have a significant impact on the outcome of health crises such as the current one. But in addition to the importance of expanding this type of research, we strongly advocate that AI innovation respects our democratic values, taking into account ethical considerations and the privacy of our citizens," said researchers Pieter Libin and Prof. Ann Nowé of the VUB Artificial Intelligence Lab.
AI initiatives in the fight against COVID-19
The VUB has set up several AI initiatives, both at university and international level. In collaboration with the medical experts of the Universitair Ziekenhuis Brussel, the Artificial Intelligence Lab set up two AI-driven projects, on the one hand around the analysis of lung scans (in collaboration with KU Leuven and EZ Tilburg) and on the other hand around the development of an AI application that can support doctors in the intensive care unit. VUB researchers are also involved in the EU-wide COVID-19 Taskforce of CLAIRE (Confederation of Laboratories of Artificial Intelligence Research in Europe) to collect and offer cross-border AI expertise and initiatives (https://covid19.claire-ai.org/).
*The research was conducted by Pieter Libin and supervised by Prof. Ann Nowé of the VUB Artificial Intelligence Lab (https://ai.vub.ac.be) and Prof. Phillippe Lemey (Rega-Institute, KU Leuven), in collaboration with Prof. Niel Hens (UHasselt and UAntwerp). It was supported by the FWO-Flanders through a strategic basic research grant (FWO-SB) and by the Flemish Government within the framework of the AI Research Programme.